Long-haul optical data links are being more extensively used for the reliable transmission of data. The high data rates and low optical attenuation associated with fiber optic links are well-established and are becoming more appreciated as the links have become more economical than alternatives based upon electrical coaxial cables. In spite of the relatively low magnitude of optical signal loss during transmission, the intrinsic linear attenuation law of lightwave energy in optical fibers, has resulted in the necessity of optical repeater nodes to amplify and/or to regenerate the digital optical bit streams in long-haul terrestrial and undersea communication systems. Typically, unrepeatered distances extend from 30 to 70 kilometers in length, depending upon the fiber loss at the selected transmission wavelength which is ordinarily 1.3 or 1.55 microns, respectively.
Repeaterless links or longer unrepeatered distances reaching from 200 to 400 kilometers, for example, would represent a less expensive and more reliable system. This feature is particularly attractive in the realm of undersea communications, since it would relieve the initial material, equipment and installation expense of undersea repeaters, and provide a reduction in the recovery and servicing expenses associated with repair of undersea optical nodes.
An innovative approach for improving an optical long-haul communications link capability might be the use of nonlinear optical effects for enhanced fiber signal transmission. This approach to date has not been obvious to those associated with this art, since fibers are usually considered as being completely passive; i.e., a more or less linear media for the transmission of optical data. Ironically, some influential papers written in the 1970's to introduce and promote nonlinear "active" properties of optical fibers, simultaneously argued that these nonlinear effects (stimulated Raman scattering, stimulated Brillouin scattering, self-phase modulation, and stimulated four-photon mixing) constituted "processes imposing limits on the peak power of fiber transmission systems", see "Nonlinear Properties of Optical Fibers", by R.H. Stolen, appearing in Optical Fiber Telecommunications, S.E. Miller and A.G. Chynoweth, Editors, Academic Press, 1979, and "Nonlinearity in Fiber Transmission", by R.H. Stolen, Proceedings IEEE, Vol. 68, No. 10, October 1980.
The misconception, or rather, misdirection of insight, that fiber nonlinearities impose limits on peak power transmission in fiber optic systems, apparently arose from an unconscious, arbitrary restriction to linear systems. In these systems and in this context, frequency conversion was misconstrued as a "power-dependent loss mechanism....producing amplitude distortion at the receiver if the detector is intrinsically frequency sensitive or narrow band filters are used"; see "Optical Power Handling Capacity of Low Loss Optical Fibers as Determined by SRS and SBS", by R.G. Smith, Journal of Applied Optics, Vol. 11, No. 11, November 1972. In view of the dominant influence of linear fiber optic systems which heretofore represent nearly the entire scope and content of the prior art in fiber communications, the above misleading notions can be excused. Embracing the much broader perspective of nonlinear light wave communications, however, provides the basis for properly introducing especially designed segments of active fiber-optic lines, and for resolution of all the difficulties associated with light transmission systems utilizing energy densities sufficiently high to induce nonlinear effects.
In spite of the initially discouraging view with respect to utilizing high power densities accompanied by nonlinear optical effects in fiber optic communication systems, laboratory workers in the late 70's, and early 80's recognized several important features of the indirect methods needed to take advantage of nonlinear optical techniques for the purpose of signal amplification. To appreciate the enhancement potential of nonlinear optical effects in fiber optic communications, certain innovative points have to be understood. The first point is that the interaction of light from a pump and signal laser, respectively, within a single fiber constitutes an essential feature of nonlinear signal amplification. The second point is that the frequency conversion of the high energy pump laser will transfer energy into the signal band which is at a lower frequency, and that the nature of this transformation can be qualitatively and quantitatively predicted. The last point is that the predicted frequency conversion must be taken into account in the design of a proper receiver/detector configuration.
Most of the early experiments utilized the nonlinear effect known as stimulated Raman scattering (SRS), in order to obtain, during propagation through an optical fiber, the frequency conversion and the transfer of energy from the pump wave into the signal wave. The development of signal amplification via SRS is summarized in the co-pending patent application for a "Tapered Fiber Amplifier" (NC 70839).
In addition, many laboratory experiments were conducted from 1974-1980 to show the feasibility of signal amplification by the nonlinear optical process of stimulated four photon mixing (SFPM). The experimentation revealed certain difficulties in obtaining a tunable pump laser configuration for amplifying signal wavelengths in the 1.1 to 1.6 micron range. For example, when a Nd:YAG laser is used as a relatively high-energy pump, the stimulated four-photon mixing (SFPM) process produces frequency conversion and amplification of a signal at approximately 1.5 microns, but appears possible only by two successive stages of parametric generation using bulk optical components. These stages included: first, light from the Nd:YAG would have to pass through a bulk crystal for second harmonic generation; and secondly, through an angle tuned crystal for further parametric generation. The complicated nature of this setup, and the intrinsic inefficiencies were considered prohibitive for practical situations [further insight is provided by "A 1.4-4.0 micron high-energy angle-tuned LiNbO.sub.3 parametric oscillator", R.L. Herbst et al, Applied Physics Letters, Vol.25, No.9, Nov. 1974, p. 520-22; also, Introduction to Optical Electronics, by A. Yariv (Holt, Rinehart & Winston, 2nd Ed., 1976, p. 222-239 )].
Stimulated four photon mixing or parametric generation can occur only by phase-matching of the pump waves to the signal and idler waves in the propagation medium. During the period from 1974 to 1980, it was determined and verified experimentally that the phase-match condition could be satisfied in certain optical fibers by tailoring their opto-geometrical parameters; see "Phase-matched three-wave mixing in silica fiber optical waveguides," R.H. Stolen, J.E. Bjorkholm, and A. Ashkin, Applied Physics Letters, Vol.24, No.7, Apr. 1974, p. 308- 10; "Optical fiber modes using stimulated four photon mixing,"R.H. Stolen and W.N. Leibolt, Applied Optics, Vol.15, No.1, Jul. 1975, pg. 239-43; and "Efficient Large-Frequency-Shifted Three-Wave Mixing in Low Dispersion Wavelength Region in Single-Mode Optical Fibre," K. Washio et al, Electronics Letters, Vol.16, No. 17, Aug. 1980, p. 658-660.
Finally, in 1981, Chinlon Lin et al. established a clear criteria for "Phase matching in the minimum-chromatic-dispersion region of single-mode fibers for stimulated four-photon mixing," Optics Letters, Vol. 6, No. 10, Oct. 1981, p. 493-495. However, signal amplification by SFPM in optical fibers has been demonstrated heretofore on precision optical benches in the laboratory only.
Thus, a continuing need exists in the state-of-the-art for a practical, field installed device for amplification of 1.3 and 1.55 micron laser diode signals by frequency conversion of a pump wave via the phenomena known as stimulated four-photon mixing, so as to increase the distance between optical repeaters and/or regenerator nodes in undersea long-haul fiber optic transmission links.